Machine learning in trading: theory, models, practice and algo-trading - page 2469
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
How is this ??
Everything you need is in the OHLC, the rest of the data is derived from it. The main thing is a flexible algorithm that will find a way to convert OHLC to the data that has more weight, again that is the task of the machine. In addition only OHLC is the same almost everywhere, if you look at tick volumes or something else this data is different, as well as spreads and so on. If you look at the chart in the terminal you'll see everything you need.
All that is needed is in the OHLC, the other data is derived from it. The main thing is a flexible algorithm that will find a way to convert OHLC to the data that has more weight, again, that is the task of the machine. In addition only OHLC is the same almost everywhere, if you look at tick volumes or something else this data is different, as well as spreads and so on. This is what you need on the chart in the terminal.
If you're relying on your own 4-year experience with model tracing, it seems to me that OHLC is not enough to make money, and a model almost never works with new data, and if it works, you can't say it happened or not... However much you crossvalidate it...
No one is interested in a chart of average positions of traders at 10 brokers?
No one is interested in the fact that the price has an inverse correlation with the mood of traders?
[1]
-0.74
Doesn't it explain why models don't work with new data, why traders are always losing, why the market mechanics itself is uncovered?
Look at TheExpert's products and his posts. He is interested in this topic and has great products in the marketplace on this topic. Ask him what he thinks about it now.
I'll add that I've been doing a bit of sentiment research too. Studied COT. The theme is that indeed sentiment is positively related to price. But when I looked at the relationship, it turns out that the range is strongly related to price, and not vice versa. Conclusions are up to you.
why models don't work on new data, why traders are constantly losing, the mechanics of the market itself reveals... no ??? not interested ???
because the market is not a model but a reality (economic reality)... because you don't have to disconnect price movements from understanding economic assumptions, reactions and consequences... much less from your eyes... because a trader must always understand which market to trade and which not, if he's a trader (from this comes the understanding of when to turn on a robot and when not)... You don't normally turn on the coffee machine to heat up a pizza...
Check out TheExpert's products and his posts. He's been interested in this topic and has great products in the marketplace on this topic. Ask him what he thinks about it now.
Yes, I remember TheExpert, I was into oanda before him...
For my own part I've been doing a bit of sentiment too. I studied COT. My point is that sentiment really is positively correlated with price. But when I looked at the correlation, it turned out that the range strongly depends on the price and not vice versa. You draw your own conclusions.
There is mathematics for that, DSP, etc..
Approximated the price and range to smooth out, checked for cross-correlation of range and price to find out "what goes where.
You're right, the price on average overrides the range, but it is on average and not always, I'll look ...
If tomorrow I don't have a hard time, I'll try some model on all 10 brokers, maybe something interesting will pop up.
Because the market is not a model, but reality (economic reality).
That's what a model is for, to be built for reality, read wikipedia or something...
because you shouldn't disconnect price movements from your understanding of economic assumptions, reactions and consequences... much less your own eyes...
words are beautiful but.... naked populism...
Because a trader must always understand what market to trade and what not to trade, if he is a trader (hence, the understanding of when to turn on the robot and when not to)
But he never understands, moreover he almost always does the opposite, the same thing with turning on / off the robot, the market state identification and other bullshit that you read in your trading blogs, the market turns against the expectations of participants with a correlation of-0.7
that's the whole story...
Man, guys, we should not argue about who started using the sanction, but about who understood it better and was able to get something useful out of it. I got acquainted with COT back in 2008 and what's the use of it.
No one needs it)) Everyone wants a grail, and even when they get it they start to argue, prove their point and don't even try to do what they say, that's the truth, plus their education is on a plateau...